def __init__(self, filename, encoding='iso8859-1'):
"""'encoding' sets the encoding assumed of log messages and
delta text in RCS files.
"""
self.filename = filename
self.encoding = encoding
self.rcsfile = rcsparse.rcsfile(filename)
import rcsparse
import md5
f = rcsparse.rcsfile("test,v")
print f.head
print f.branch
s = f.symbols
print s["RELENG_4"]
print s.items()
r = f.revs
i = r.items()
print i
print f.getlog(f.sym2rev("RELENG_4"))
def expand_keyword(self, filename, r):
rcs = rcsparse.rcsfile(filename)
rev = rcs.revs[r]
mode = self.kflag_get(rcs.expand)
if (mode & (self.RCS_KWEXP_NONE | self.RCS_KWEXP_OLD)) != 0:
return rcs.checkout(rev[0])
s = logbuf = ''
for line in rcs.checkout(rev[0]).split('\n'):
while True:
m = self.re_kw.match(line)
if m is None:
break
if len(line) > m.end(1) and line[m.end(1)] == '$':
dsign = m.end(1)
else:
try:
dsign = string.index(line, '$', m.end(1))
if dsign < 0:
break
except:
break
prefix = line[:m.start(1)-1]
line = line[dsign + 1:]
s += prefix
expbuf = ''
if (mode & self.RCS_KWEXP_NAME) != 0:
expbuf += '$'
expbuf += m.group(1)
if (mode & self.RCS_KWEXP_VAL) != 0:
expbuf += ': '
if (mode & self.RCS_KWEXP_VAL) != 0:
expkw = self.rcs_expkw[m.group(1)]
if (expkw & self.RCS_KW_RCSFILE) != 0:
expbuf += filename \
if (expkw & self.RCS_KW_FULLPATH) != 0 \
else os.path.basename(filename)
expbuf += " "
if (expkw & self.RCS_KW_REVISION) != 0:
expbuf += rev[0]
expbuf += " "
if (expkw & self.RCS_KW_DATE) != 0:
expbuf += time.strftime("%Y/%m/%d %H:%M:%S ", \
time.gmtime(rev[1]))
if (expkw & self.RCS_KW_MDOCDATE) != 0:
d = time.gmtime(rev[1])
expbuf += time.strftime("%B%e %Y " \
if (d.tm_mday < 10) else "%B %e %Y ", d)
if (expkw & self.RCS_KW_AUTHOR) != 0:
expbuf += rev[2]
expbuf += " "
if (expkw & self.RCS_KW_STATE) != 0:
expbuf += rev[3]
expbuf += " "
if (expkw & self.RCS_KW_LOG) != 0:
p = prefix
expbuf += filename \
if (expkw & self.RCS_KW_FULLPATH) != 0 \
else os.path.basename(filename)
expbuf += " "
logbuf += '%sRevision %s ' % (p, rev[0])
logbuf += time.strftime("%Y/%m/%d %H:%M:%S ",\
time.gmtime(rev[1]))
logbuf += rev[2] + '\n'
for lline in rcs.getlog(rev[0]).rstrip().split('\n'):
if lline == '':
logbuf += p.rstrip() + '\n'
else:
logbuf += p + lline.lstrip() + '\n'
if line == '':
logbuf += p.rstrip() + '\n'
else:
logbuf += p + line.lstrip() + '\n'
line = ''
if (expkw & self.RCS_KW_SOURCE) != 0:
expbuf += filename
expbuf += " "
if (expkw & (self.RCS_KW_NAME | self.RCS_KW_LOCKER)) != 0:
expbuf += " "
if (mode & self.RCS_KWEXP_NAME) != 0:
expbuf += '$'
s += expbuf[:255]
s += line + '\n'
if len(logbuf) > 0:
s += logbuf
logbuf = ''
return s[:-1]
import rcsparse
f = rcsparse.rcsfile('test,v')
print(f.head)
print(f.branch)
s = f.symbols
print(s['RELENG_4'])
print(list(s.items()))
r = f.revs
i = list(r.items())
print(i)
print(f.getlog(f.sym2rev('RELENG_4')).decode('ascii'))
print('1.1' in r)
for tag in k.tags:
extags.add(tag)
continue
if k.max_time > max_time_max:
break
marks = {}
for f in k.revs:
if not do_incremental:
marks[f.markseq] = f
else:
markseq = markseq + 1
git_dump_file(f.path, f.rev, rcs, markseq)
marks[markseq] = f
log = rcsparse.rcsfile(k.revs[0].path).getlog(k.revs[0].rev)
for i, e in enumerate(log_encodings):
try:
how = 'ignore' if i == len(log_encodings) - 1 else 'strict';
log = log.decode(e, how)
break
except:
pass
log = log.encode('utf-8', 'ignore')
print 'commit refs/heads/' + git_branch
markseq = markseq + 1
print 'mark :%d' % (markseq)
email = k.author if email_domain is None \
else k.author + '@' + email_domain
print 'author %s <%s> %d +0000' % (k.author, email, k.min_time)
def parse_file(self, path):
rtags = dict()
rcsfile=rcsparse.rcsfile(path)
path_related = path[len(self.cvsroot) + 1:][:-2]
branches = {'1': 'HEAD', '1.1.1': 'VENDOR' }
have_111 = False
for k,v in rcsfile.symbols.items():
r = v.split('.')
if len(r) == 3:
branches[v] = 'VENDOR'
elif len(r) >= 3 and r[-2] == '0':
z = reduce(lambda a, b: a + '.' + b, r[:-2] + r[-1:])
branches[reduce(lambda a, b: a + '.' + b, r[:-2] + r[-1:])] = k
if len(r) == 2 and branches[r[0]] == 'HEAD':
if not rtags.has_key(v):
rtags[v] = list()
rtags[v].append(k)
# sort by time and revision
revs = sorted(rcsfile.revs.items(), \
lambda a,b: cmp(a[1][1], b[1][1]) or cmp(b[1][0], a[1][0]))
p = '0'
novendor = False
have_initial_revision = False
last_vendor_status = None
for k,v in revs:
r = k.split('.')
if len(r) == 4 and r[0] == '1' and r[1] == '1' and r[2] == '1' \
and r[3] == '1':
if have_initial_revision:
continue
if v[3] == 'dead':
continue
last_vendor_status = v[3]
have_initial_revision = True
elif len(r) == 4 and r[0] == '1' and r[1] == '1' and r[2] == '1':
if novendor:
continue
last_vendor_status = v[3]
elif len(r) == 2:
if r[0] == '1' and r[1] == '1':
if have_initial_revision:
continue
if v[3] == 'dead':
continue
have_initial_revision = True
elif r[0] == '1' and r[1] != '1':
novendor = True
if last_vendor_status == 'dead' and v[3] == 'dead':
last_vendor_status = None
continue
last_vendor_status = None
else:
# trunk only
continue
if self.dumpfile:
self.markseq = self.markseq + 1
git_dump_file(path, k, self.rcs, self.markseq)
b = reduce(lambda a, b: a + '.' + b, r[:-1])
try:
a = ChangeSetKey(branches[b], v[2], v[1], rcsfile.getlog(v[0]),
v[6], self.fuzzsec)
except Exception as e:
print >>sys.stderr, 'Aborted at %s %s' % (path, v[0])
raise e
a.put_file(path, k, v[3], self.markseq)
while self.changesets.has_key(a):
c = self.changesets[a]
del self.changesets[a]
c.merge(a)
a = c
self.changesets[a] = a
p = k
if rtags.has_key(k):
for t in rtags[k]:
if not self.tags.has_key(t) or \
self.tags[t].max_time < a.max_time:
self.tags[t] = a
def expand_keyword(self, filename, r):
rcs = rcsparse.rcsfile(filename)
rev = rcs.revs[r]
mode = self.kflag_get(rcs.expand)
if (mode & (self.RCS_KWEXP_NONE | self.RCS_KWEXP_OLD)) != 0:
return rcs.checkout(rev[0])
s = logbuf = ''
for line in rcs.checkout(rev[0]).split('\n'):
while True:
m = self.re_kw.match(line)
if m is None:
break
if len(line) > m.end(1) and line[m.end(1)] == '$':
dsign = m.end(1)
else:
try:
dsign = string.index(line, '$', m.end(1))
if dsign < 0:
break
except:
break
prefix = line[:m.start(1) - 1]
line = line[dsign + 1:]
s += prefix
expbuf = ''
if (mode & self.RCS_KWEXP_NAME) != 0:
expbuf += '$'
expbuf += m.group(1)
if (mode & self.RCS_KWEXP_VAL) != 0:
expbuf += ': '
if (mode & self.RCS_KWEXP_VAL) != 0:
expkw = self.rcs_expkw[m.group(1)]
if (expkw & self.RCS_KW_RCSFILE) != 0:
expbuf += filename \
if (expkw & self.RCS_KW_FULLPATH) != 0 \
else os.path.basename(filename)
expbuf += " "
if (expkw & self.RCS_KW_REVISION) != 0:
expbuf += rev[0]
expbuf += " "
if (expkw & self.RCS_KW_DATE) != 0:
expbuf += time.strftime("%Y/%m/%d %H:%M:%S ", \
time.gmtime(rev[1]))
if (expkw & self.RCS_KW_MDOCDATE) != 0:
d = time.gmtime(rev[1])
expbuf += time.strftime("%B%e %Y " \
if (d.tm_mday < 10) else "%B %e %Y ", d)
if (expkw & self.RCS_KW_AUTHOR) != 0:
expbuf += rev[2]
expbuf += " "
if (expkw & self.RCS_KW_STATE) != 0:
expbuf += rev[3]
expbuf += " "
if (expkw & self.RCS_KW_LOG) != 0:
p = prefix
expbuf += filename \
if (expkw & self.RCS_KW_FULLPATH) != 0 \
else os.path.basename(filename)
expbuf += " "
logbuf += '%sRevision %s ' % (p, rev[0])
logbuf += time.strftime("%Y/%m/%d %H:%M:%S ",\
time.gmtime(rev[1]))
logbuf += rev[2] + '\n'
for lline in rcs.getlog(rev[0]).rstrip().split('\n'):
if lline == '':
logbuf += p.rstrip() + '\n'
else:
logbuf += p + lline.lstrip() + '\n'
if line == '':
logbuf += p.rstrip() + '\n'
else:
logbuf += p + line.lstrip() + '\n'
line = ''
if (expkw & self.RCS_KW_SOURCE) != 0:
expbuf += filename
expbuf += " "
if (expkw & (self.RCS_KW_NAME | self.RCS_KW_LOCKER)) != 0:
expbuf += " "
if (mode & self.RCS_KWEXP_NAME) != 0:
expbuf += '$'
s += expbuf[:255]
s += line + '\n'
if len(logbuf) > 0:
s += logbuf
logbuf = ''
return s[:-1]
def parse_file(self, path):
rtags = dict()
rcsfile = rcsparse.rcsfile(path)
path_related = path[len(self.cvsroot) + 1:][:-2]
branches = {'1': 'HEAD', '1.1.1': 'VENDOR'}
have_111 = False
for k, v in rcsfile.symbols.items():
r = v.split('.')
if len(r) == 3:
branches[v] = 'VENDOR'
elif len(r) >= 3 and r[-2] == '0':
z = reduce(lambda a, b: a + '.' + b, r[:-2] + r[-1:])
branches[reduce(lambda a, b: a + '.' + b, r[:-2] + r[-1:])] = k
if len(r) == 2 and branches[r[0]] == 'HEAD':
if not rtags.has_key(v):
rtags[v] = list()
rtags[v].append(k)
# sort by time and revision
revs = sorted(rcsfile.revs.items(), \
lambda a,b: cmp(a[1][1], b[1][1]) or cmp(b[1][0], a[1][0]))
p = '0'
novendor = False
have_initial_revision = False
last_vendor_status = None
for k, v in revs:
r = k.split('.')
if len(r) == 4 and r[0] == '1' and r[1] == '1' and r[2] == '1' \
and r[3] == '1':
if have_initial_revision:
continue
if v[3] == 'dead':
continue
last_vendor_status = v[3]
have_initial_revision = True
elif len(r) == 4 and r[0] == '1' and r[1] == '1' and r[2] == '1':
if novendor:
continue
last_vendor_status = v[3]
elif len(r) == 2:
if r[0] == '1' and r[1] == '1':
if have_initial_revision:
continue
if v[3] == 'dead':
continue
have_initial_revision = True
elif r[0] == '1' and r[1] != '1':
novendor = True
if last_vendor_status == 'dead' and v[3] == 'dead':
last_vendor_status = None
continue
last_vendor_status = None
else:
# trunk only
continue
if self.dumpfile:
self.markseq = self.markseq + 1
git_dump_file(path, k, self.rcs, self.markseq)
b = reduce(lambda a, b: a + '.' + b, r[:-1])
try:
a = ChangeSetKey(branches[b], v[2], v[1], rcsfile.getlog(v[0]),
v[6], self.fuzzsec)
except Exception as e:
print >> sys.stderr, 'Aborted at %s %s' % (path, v[0])
raise e
a.put_file(path, k, v[3], self.markseq)
while self.changesets.has_key(a):
c = self.changesets[a]
del self.changesets[a]
c.merge(a)
a = c
self.changesets[a] = a
p = k
if rtags.has_key(k):
for t in rtags[k]:
if not self.tags.has_key(t) or \
self.tags[t].max_time < a.max_time:
self.tags[t] = a
for tag in k.tags:
extags.add(tag)
continue
if k.max_time > max_time_max:
break
marks = {}
for f in k.revs:
if not do_incremental:
marks[f.markseq] = f
else:
markseq = markseq + 1
git_dump_file(f.path, f.rev, rcs, markseq)
marks[markseq] = f
log = rcsparse.rcsfile(k.revs[0].path).getlog(k.revs[0].rev)
for i, e in enumerate(log_encodings):
try:
how = 'ignore' if i == len(log_encodings) - 1 else 'strict'
log = log.decode(e, how)
break
except:
pass
log = log.encode('utf-8', 'ignore')
print 'commit refs/heads/' + git_branch
markseq = markseq + 1
print 'mark :%d' % (markseq)
email = k.author if email_domain is None \
else k.author + '@' + email_domain
print 'author %s <%s> %d +0000' % (k.author, email, k.min_time)
found_last_revision = False
for i, k in enumerate(changesets):
if do_incremental and not found_last_revision:
if k.min_time == svn.last_date and k.author == last_author:
found_last_revision = True
continue
if k.max_time > max_time_max:
break
if not printOnce:
print 'SVN-fs-dump-format-version: 2'
print ''
printOnce = True
# parse the first file to get log
finfo = k.revs[0]
rcsfile = rcsparse.rcsfile(finfo.path)
log = rcsparse.rcsfile(k.revs[0].path).getlog(k.revs[0].rev)
for i, e in enumerate(log_encodings):
try:
how = 'ignore' if i == len(log_encodings) - 1 else 'strict'
log = log.decode(e, how)
break
except:
pass
log = log.encode('utf-8', 'ignore')
if email_domain is None:
email = k.author
else:
email = k.author + '@' + email_domain
found_last_revision = False
for i, k in enumerate(changesets):
if do_incremental and not found_last_revision:
if k.min_time == svn.last_date and k.author == last_author:
found_last_revision = True
continue
if k.max_time > max_time_max:
break
if not printOnce:
print 'SVN-fs-dump-format-version: 2'
print ''
printOnce = True
# parse the first file to get log
finfo = k.revs[0]
rcsfile = rcsparse.rcsfile(finfo.path)
log = rcsparse.rcsfile(k.revs[0].path).getlog(k.revs[0].rev)
for i, e in enumerate(log_encodings):
try:
how = 'ignore' if i == len(log_encodings) - 1 else 'strict';
log = log.decode(e, how)
break
except:
pass
log = log.encode('utf-8', 'ignore')
if email_domain is None:
email = k.author
else:
email = k.author + '@' + email_domain